FreeRDP winpr-makecert CVE-2026-109244

Exploitation requires that a private key or PFX file was generated on the host using FreeRDP's winpr-makecert utility (versions 2.0.0 through 3.31.x), that this file resides on a multi-user Unix system, and that the attacker has an existing local low-privileged shell account (PR:L). The flaw is the key being written without owner-only (e.g. 0600) permissions, making it world-readable, and/or the PFX being protected only by winpr-makecert's hard-coded default password. Limiting factors: no remote or network path (AV:L), the victim must actively have used winpr-makecert to produce the key material (UI:P), and single-user or properly-umask'd hosts where no such keys exist are not exploitable - this does not affect default FreeRDP client/server RDP sessions that never invoke the certificate tool.

This is a genuinely medium-severity, not a high-priority, issue, and the signals are internally consistent rather than conflicting. The vendor-assigned CVSS 4.0 vector (AV:L/AC:L/AT:N/PR:L/UI:P) confirms the flaw is exploitable only by a local, already-authenticated low-privileged user (PR:L) on a shared host, with some passive user interaction (UI:P) - there is no network reachability. No EPSS score, CISA KEV listing, or public proof-of-concept is present in the provided intelligence, so there is no public exploit identified at time of analysis and no evidence of active exploitation. The combination of a local-only attack surface, the requirement that a victim actually used winpr-makecert to produce keys, and the absence of any exploitation telemetry means this should be remediated through normal patch cycles rather than treated as an emergency. The high confidentiality and integrity impact (VC:H/VI:H) reflects the value of a stolen private key - it enables impersonation/MITM - but that payoff only materializes on multi-user Unix hosts where such keys exist and are left world-readable.

On a shared Unix server, an administrator generates TLS certificates with FreeRDP's winpr-makecert, which writes the unencrypted private key (or a PFX protected only by a hard-coded default password) without owner-only file permissions. A separate low-privileged local user simply reads the world-readable key file, or opens the PFX with the known default password, and exfiltrates the private key to impersonate the service or decrypt traffic. No public proof-of-concept is identified at time of analysis.